use "D:\Dropbox\Surge_rhetoric\Vaccines\Vaccine submission\KK_vaccines.dta", clear
gen vax_accept = vaxord
recode vax_accept (7=1) (6=1) (5=1) (4=0) (3=0) (2=0) (1=0)

* Figure 1 *
* Blacks *
reg vax_accept i.efficacy_n##i.black i.duration_n##i.black i.majorside_n##i.black i.minorside_n##i.black i.fda_n##i.black i.origin_n##i.black i.endorsed_n##i.black if black==1 | white == 1, cluster(respondent) coeflegend
margins efficacy_n duration_n majorside_n minorside_n  fda_n origin_n endorsed_n, at(black=1) post
estimates store black
reg vax_accept i.efficacy_n##i.black i.duration_n##i.black i.majorside_n##i.black i.minorside_n##i.black i.fda_n##i.black i.origin_n##i.black i.endorsed_n##i.black, cluster(respondent) coeflegend
margins efficacy_n duration_n majorside_n minorside_n  fda_n origin_n endorsed_n, at(black=0) post
estimates store whites
coefplot  whites black, drop(_cons) omitted baselevels  ///
headings(1.efficacy_n = "{bf:Efficacy}" ///  
1.duration_n = "{bf:Protection Duration}" ///            
1.majorside_n = "{bf:Major Side Effects}" ///
1.minorside_n = "{bf:Minor Side Effects}" ///
1.fda_n = "{bf:FDA Approval}" ///
1.origin_n = "{bf:Origin}" ///
1.endorsed_n = "{bf:Endorsement}" , labsize(vsmall)) graphregion(color(white)) ylab(, labs(vsmall)) ///
xti("Marginal Mean of Taking Vaccine", size(small)) level(95) xlabel(.4 .45 .5 .55 .6 .65 .7)  mlcolor(black) mcolor(black) msize(medsmall) ti("Blacks") legend(order(4 "Blacks" 2 "Whites") size(small)) saving(black.gph, replace)
gr_edit plotregion1.plot4.style.editstyle marker(symbol(Sh)) editcopy 

* Model significance test *
reg vax_accept i.efficacy_n i.duration_n i.majorside_n i.minorside_n i.fda_n i.origin_n i.endorsed_n if white == 1|black==1 
est store white
reg vax_accept i.efficacy_n##i.black i.duration_n##i.black i.majorside_n##i.black i.minorside_n##i.black i.fda_n##i.black i.origin_n##i.black i.endorsed_n##i.black if white==1|black==1
est store black
ftest white black

pwcompare i.efficacy_n#black, group level(95)
pwcompare i.duration_n#black, group level(95)
pwcompare i.majorside_n#black, group level(95)
pwcompare i.minorside_n#black, group level(95)
pwcompare i.fda_n#black, group level(95)
pwcompare i.origin_n#black, group level(95)
pwcompare i.endorsed_n#black, group level(95)

* Illustration of how significance depends on choice of baseline -- 
reg vax_accept i.efficacy_n##i.black i.duration_n##i.black i.majorside_n##i.black i.minorside_n##i.black i.fda_n##i.black i.origin_n##i.black i.endorsed_n##i.black if white==1|black==1
gen endorsed1xblack = endorsed1*black
gen endorsed2xblack = endorsed2*black
gen endorsed3xblack = endorsed3*black
gen endorsed4xblack = endorsed4*black

reg vax_accept i.efficacy_n##i.black i.duration_n##i.black i.majorside_n##i.black i.minorside_n##i.black i.fda_n##i.black i.origin_n##i.black endorsed2 endorsed3 endorsed4 endorsed2xblack endorsed3xblack endorsed4xblack if white==1|black==1
* Now make Biden the omitted baseline *
reg vax_accept i.efficacy_n##i.black i.duration_n##i.black i.majorside_n##i.black i.minorside_n##i.black i.fda_n##i.black i.origin_n##i.black endorsed2 endorsed1 endorsed4 endorsed2xblack endorsed1xblack endorsed4xblack if white==1|black==1

* Latinos *
reg vax_accept i.efficacy_n##i.latino i.duration_n##i.latino i.majorside_n##i.latino i.minorside_n##i.latino i.fda_n##i.latino i.origin_n##i.latino i.endorsed_n##i.latino if latino==1 | white == 1, cluster(respondent) coeflegend
margins efficacy_n duration_n majorside_n minorside_n  fda_n origin_n endorsed_n, at(latino=1) post
estimates store latino
reg vax_accept i.efficacy_n##i.latino i.duration_n##i.latino i.majorside_n##i.latino i.minorside_n##i.latino i.fda_n##i.latino i.origin_n##i.latino i.endorsed_n##i.latino, cluster(respondent) coeflegend
margins efficacy_n duration_n majorside_n minorside_n  fda_n origin_n endorsed_n, at(latino=0) post
estimates store whites
coefplot  whites latino, drop(_cons) omitted baselevels  ///
headings(1.efficacy_n = "{bf:Efficacy}" ///  
1.duration_n = "{bf:Protection Duration}" ///            
1.majorside_n = "{bf:Major Side Effects}" ///
1.minorside_n = "{bf:Minor Side Effects}" ///
1.fda_n = "{bf:FDA Approval}" ///
1.origin_n = "{bf:Origin}" ///
1.endorsed_n = "{bf:Endorsement}" , labsize(vsmall)) graphregion(color(white)) ylab(, labs(vsmall)) ///
xti("Marginal Mean of Taking Vaccine", size(small)) level(95) xlabel(.4 .45 .5 .55 .6 .65 .7)  mlcolor(latino) mcolor(black) msize(medsmall) ti("Latinx") legend(order(4 "Latinx" 2 "Whites") size(small)) saving(latino.gph, replace)
gr_edit plotregion1.plot4.style.editstyle marker(symbol(Sh)) editcopy

* Model significance test *
reg vax_accept i.efficacy_n i.duration_n i.majorside_n i.minorside_n i.fda_n i.origin_n i.endorsed_n  
est store white
reg vax_accept i.efficacy_n##i.latino i.duration_n##i.latino i.majorside_n##i.latino i.minorside_n##i.latino i.fda_n##i.latino i.origin_n##i.latino i.endorsed_n##i.latino
est store latino
ftest white latino

pwcompare i.efficacy_n#latino, group level(95)
pwcompare i.duration_n#latino, group level(95)
pwcompare i.majorside_n#latino, group level(95)
pwcompare i.minorside_n#latino, group level(95)
pwcompare i.fda_n#latino, group level(95)
pwcompare i.origin_n#latino, group level(95)
pwcompare i.endorsed_n#latino, group level(95)

graph combine black.gph latino.gph, graphregion(color(white)) 
gr_edit style.editstyle boxstyle(shadestyle(color(white))) editcopy
gr_edit plotregion1.graph1.plotregion1.plot4.style.editstyle marker(symbol(Sh)) editcopy
gr_edit plotregion1.graph2.plotregion1.plot4.style.editstyle marker(symbol(Sh)) editcopy
gr_edit legend.plotregion1.key[1].view.style.editstyle marker(symbol(Sh)) editcopy

* SI Table 2 *
reg vax_accept i.efficacy_n##i.black i.duration_n##i.black i.majorside_n##i.black i.minorside_n##i.black i.fda_n##i.black i.origin_n##i.black i.endorsed_n##i.black if black==1 | white == 1, cluster(respondent) coeflegend
outreg2 using sitable1model1, word keep(1.black 2.efficacy_n  3.efficacy_n 2.efficacy_n#1.black 3.efficacy_n#1.black 2.duration_n 2.duration_n#1.black ///
 2.majorside_n 2.majorside_n#1.black 2.minorside_n 2.minorside_n#1.black 2.fda_n 2.fda_n#1.black 2.origin_n 3.origin_n ///
 2.origin_n#1.black 3.origin_n#1.black 2.endorsed_n 3.endorsed_n 4.endorsed_n 2.endorsed_n#1.black 3.endorsed_n#1.black 4.endorsed_n#1.black) ///
 sortvar(1.black  2.efficacy_n  3.efficacy_n ) replace  dec(2) 
reg vax_accept i.efficacy_n##i.latino i.duration_n##i.latino i.majorside_n##i.latino i.minorside_n##i.latino i.fda_n##i.latino i.origin_n##i.latino i.endorsed_n##i.latino if latino==1 | white == 1, cluster(respondent) coeflegend
outreg2 using sitable1model2, word keep(1.latino 2.efficacy_n  3.efficacy_n 2.efficacy_n#1.latino 3.efficacy_n#1.latino 2.duration_n 2.duration_n#1.latino ///
 2.majorside_n 2.majorside_n#1.latino 2.minorside_n 2.minorside_n#1.latino 2.fda_n 2.fda_n#1.latino 2.origin_n 3.origin_n ///
 2.origin_n#1.latino 3.origin_n#1.latino 2.endorsed_n 3.endorsed_n 4.endorsed_n 2.endorsed_n#1.latino 3.endorsed_n#1.latino 4.endorsed_n#1.latino) ///
 sortvar(1.latino  2.efficacy_n  3.efficacy_n ) replace  dec(2) 

 * Figure 2: Age *
gen sixtyplus = 0
replace sixtyplus = 1 if age>=60

reg vax_accept i.efficacy_n##i.sixtyplus i.duration_n##i.sixtyplus i.majorside_n##i.sixtyplus i.minorside_n##i.sixtyplus i.fda_n##i.sixtyplus i.origin_n##i.sixtyplus i.endorsed_n##i.sixtyplus, cluster(respondent) coeflegend
margins efficacy_n duration_n majorside_n minorside_n  fda_n origin_n endorsed_n, at(sixtyplus=1) post
estimates store sixtyplus
reg vax_accept i.efficacy_n##i.sixtyplus i.duration_n##i.sixtyplus i.majorside_n##i.sixtyplus i.minorside_n##i.sixtyplus i.fda_n##i.sixtyplus i.origin_n##i.sixtyplus i.endorsed_n##i.sixtyplus, cluster(respondent) coeflegend
margins efficacy_n duration_n majorside_n minorside_n  fda_n origin_n endorsed_n, at(sixtyplus=0) post
estimates store whites
coefplot  whites sixtyplus, drop(_cons) omitted baselevels  ///
headings(1.efficacy_n = "{bf:Efficacy}" ///  
1.duration_n = "{bf:Protection Duration}" ///            
1.majorside_n = "{bf:Major Side Effects}" ///
1.minorside_n = "{bf:Minor Side Effects}" ///
1.fda_n = "{bf:FDA Approval}" ///
1.origin_n = "{bf:Origin}" ///
1.endorsed_n = "{bf:Endorsement}" , labsize(vsmall)) graphregion(color(white)) ylab(, labs(vsmall)) ///
xti("Marginal Mean of Taking Vaccine", size(small)) level(95) xlabel(.4 .45 .5 .55 .6 .65 .7)  mlcolor(poc) mcolor(poc) msize(medsmall) legend(order(4 "60 plus" 2 "Under 60") size(small)) saving(race.gph, replace)
gr_edit plotregion1.plot4.style.editstyle marker(symbol(Sh)) editcopy

* Model significance test *
reg vax_accept i.efficacy_n i.duration_n i.majorside_n i.minorside_n i.fda_n i.origin_n i.endorsed_n  
est store white
reg vax_accept i.efficacy_n##i.sixtyplus i.duration_n##i.sixtyplus i.majorside_n##i.sixtyplus i.minorside_n##i.sixtyplus i.fda_n##i.sixtyplus i.origin_n##i.sixtyplus i.endorsed_n##i.sixtyplus
est store sixtyplus
ftest white sixtyplus

pwcompare i.efficacy_n#sixtyplus, group level(95)
pwcompare i.duration_n#sixtyplus, group level(95)
pwcompare i.majorside_n#sixtyplus, group level(95)
pwcompare i.minorside_n#sixtyplus, group level(95)
pwcompare i.fda_n#sixtyplus, group level(95)
pwcompare i.origin_n#sixtyplus, group level(95)
pwcompare i.endorsed_n#sixtyplus, group level(95)

* SI Table 3 *
reg vax_accept i.efficacy_n##i.sixtyplus i.duration_n##i.sixtyplus i.majorside_n##i.sixtyplus i.minorside_n##i.sixtyplus i.fda_n##i.sixtyplus i.origin_n##i.sixtyplus i.endorsed_n##i.sixtyplus, cluster(respondent) coeflegend
outreg2 using sitable2, word keep(1.sixtyplus 2.efficacy_n  3.efficacy_n 2.efficacy_n#1.sixtyplus 3.efficacy_n#1.sixtyplus 2.duration_n 2.duration_n#1.sixtyplus ///
 2.majorside_n 2.majorside_n#1.sixtyplus 2.minorside_n 2.minorside_n#1.sixtyplus 2.fda_n 2.fda_n#1.sixtyplus 2.origin_n 3.origin_n ///
 2.origin_n#1.sixtyplus 3.origin_n#1.sixtyplus 2.endorsed_n 3.endorsed_n 4.endorsed_n 2.endorsed_n#1.sixtyplus 3.endorsed_n#1.sixtyplus 4.endorsed_n#1.sixtyplus) ///
 sortvar(1.sixtyplus  2.efficacy_n  3.efficacy_n ) replace  dec(2) 

 * Figure 3: Gender *
reg vax_accept i.efficacy_n##i.female i.duration_n##i.female i.majorside_n##i.female i.minorside_n##i.female i.fda_n##i.female i.origin_n##i.female i.endorsed_n##i.female, cluster(respondent) coeflegend
margins efficacy_n duration_n majorside_n minorside_n  fda_n origin_n endorsed_n, at(female=1) post
estimates store female
reg vax_accept i.efficacy_n##i.female i.duration_n##i.female i.majorside_n##i.female i.minorside_n##i.female i.fda_n##i.female i.origin_n##i.female i.endorsed_n##i.female, cluster(respondent) coeflegend
margins efficacy_n duration_n majorside_n minorside_n  fda_n origin_n endorsed_n, at(female=0) post
estimates store whites
coefplot  whites female, drop(_cons) omitted baselevels  ///
headings(1.efficacy_n = "{bf:Efficacy}" ///  
1.duration_n = "{bf:Protection Duration}" ///            
1.majorside_n = "{bf:Major Side Effects}" ///
1.minorside_n = "{bf:Minor Side Effects}" ///
1.fda_n = "{bf:FDA Approval}" ///
1.origin_n = "{bf:Origin}" ///
1.endorsed_n = "{bf:Endorsement}" , labsize(vsmall)) graphregion(color(white)) ylab(, labs(vsmall)) ///
xti("Marginal Mean of Taking Vaccine", size(small)) level(95) xlabel(.4 .45 .5 .55 .6 .65 .7)  mlcolor(poc) mcolor(poc) msize(medsmall) legend(order(4 "Women" 2 "Men") size(small)) saving(race.gph, replace)
gr_edit plotregion1.plot4.style.editstyle marker(symbol(Sh)) editcopy

* Model significance test *
reg vax_accept i.efficacy_n i.duration_n i.majorside_n i.minorside_n i.fda_n i.origin_n i.endorsed_n   
est store white
reg vax_accept i.efficacy_n##i.female i.duration_n##i.female i.majorside_n##i.female i.minorside_n##i.female i.fda_n##i.female i.origin_n##i.female i.endorsed_n##i.female  
est store female
ftest white female

pwcompare i.efficacy_n#female, group level(95)
pwcompare i.duration_n#female, group level(95)
pwcompare i.majorside_n#female, group level(95)
pwcompare i.minorside_n#female, group level(95)
pwcompare i.fda_n#female, group level(95)
pwcompare i.origin_n#female, group level(95)
pwcompare i.endorsed_n#female, group level(95)

* SI Table 4 *
reg vax_accept i.efficacy_n##i.female i.duration_n##i.female i.majorside_n##i.female i.minorside_n##i.female i.fda_n##i.female i.origin_n##i.female i.endorsed_n##i.female, cluster(respondent) coeflegend
outreg2 using sitable3, word keep(1.female 2.efficacy_n  3.efficacy_n 2.efficacy_n#1.female 3.efficacy_n#1.female 2.duration_n 2.duration_n#1.female ///
 2.majorside_n 2.majorside_n#1.female 2.minorside_n 2.minorside_n#1.female 2.fda_n 2.fda_n#1.female 2.origin_n 3.origin_n ///
 2.origin_n#1.female 3.origin_n#1.female 2.endorsed_n 3.endorsed_n 4.endorsed_n 2.endorsed_n#1.female 3.endorsed_n#1.female 4.endorsed_n#1.female) ///
 sortvar(1.female  2.efficacy_n  3.efficacy_n ) replace  dec(2) 

 * Figure 4: Partisanship *
reg vax_accept i.efficacy_n##i.gop5 i.duration_n##i.gop5 i.majorside_n##i.gop5 i.minorside_n##i.gop5 i.fda_n##i.gop5 i.origin_n##i.gop5 i.endorsed_n##i.gop5  if dem5 ==1|gop5 == 1, cluster(respondent) coeflegend
margins efficacy_n duration_n majorside_n minorside_n  fda_n origin_n endorsed_n, at(gop5=1) post
estimates store gop5
reg vax_accept i.efficacy_n##i.gop5 i.duration_n##i.gop5 i.majorside_n##i.gop5 i.minorside_n##i.gop5 i.fda_n##i.gop5 i.origin_n##i.gop5 i.endorsed_n##i.gop5  if dem5 ==1|gop5 == 1, cluster(respondent) coeflegend
margins efficacy_n duration_n majorside_n minorside_n  fda_n origin_n endorsed_n, at(gop5=0) post
estimates store dems
coefplot  dems gop5, drop(_cons) omitted baselevels  ///
headings(1.efficacy_n = "{bf:Efficacy}" ///  
1.duration_n = "{bf:Protection Duration}" ///            
1.majorside_n = "{bf:Major Side Effects}" ///
1.minorside_n = "{bf:Minor Side Effects}" ///
1.fda_n = "{bf:FDA Approval}" ///
1.origin_n = "{bf:Origin}" ///
1.endorsed_n = "{bf:Endorsement}" , labsize(vsmall)) graphregion(color(white)) ylab(, labs(vsmall)) ///
xti("Marginal Mean of Taking Vaccine", size(small)) level(95) xlabel(.4 .45 .5 .55 .6 .65 .7)  mlcolor(poc) mcolor(poc) msize(medsmall) legend(order(4 "Republicans" 2 "Democrats") size(small)) saving(race.gph, replace)
gr_edit plotregion1.plot4.style.editstyle marker(symbol(Sh)) editcopy

* Model significance test *
reg vax_accept i.efficacy_n i.duration_n i.majorside_n i.minorside_n i.fda_n i.origin_n i.endorsed_n   if dem5 ==1|gop5 == 1
est store white
reg vax_accept i.efficacy_n##i.gop5 i.duration_n##i.gop5 i.majorside_n##i.gop5 i.minorside_n##i.gop5 i.fda_n##i.gop5 i.origin_n##i.gop5 i.endorsed_n##i.gop5  if dem5 ==1|gop5 == 1
est store gop5
ftest white gop5

pwcompare i.efficacy_n#gop5, group level(95)
pwcompare i.duration_n#gop5, group level(95)
pwcompare i.majorside_n#gop5, group level(95)
pwcompare i.minorside_n#gop5, group level(95)
pwcompare i.fda_n#gop5, group level(95)
pwcompare i.origin_n#gop5, group level(95)
pwcompare i.endorsed_n#gop5, group level(95)

* SI Table 5 *
reg vax_accept i.efficacy_n##i.gop5 i.duration_n##i.gop5 i.majorside_n##i.gop5 i.minorside_n##i.gop5 i.fda_n##i.gop5 i.origin_n##i.gop5 i.endorsed_n##i.gop5 if gop5==1 | dem5 == 1, cluster(respondent) coeflegend
outreg2 using sitable4, word keep(1.gop5 2.efficacy_n  3.efficacy_n 2.efficacy_n#1.gop5 3.efficacy_n#1.gop5 2.duration_n 2.duration_n#1.gop5 ///
 2.majorside_n 2.majorside_n#1.gop5 2.minorside_n 2.minorside_n#1.gop5 2.fda_n 2.fda_n#1.gop5 2.origin_n 3.origin_n ///
 2.origin_n#1.gop5 3.origin_n#1.gop5 2.endorsed_n 3.endorsed_n 4.endorsed_n 2.endorsed_n#1.gop5 3.endorsed_n#1.gop5 4.endorsed_n#1.gop5) ///
 sortvar(1.gop5  2.efficacy_n  3.efficacy_n ) replace  dec(2) 

 * Table 2 *
 * Note -- these questions were only asked once and so are the same for each choice_set *
ologit flu_vaccine dem5 gop5 black latino female age education if choice_set == 11
outreg2 using sitable2, word dec(2) replace
ologit vax_safety dem5 gop5 black latino female age education knowcovidsevere if choice_set == 11
outreg2 using sitable2, word dec(2) append
logit knowcovidsevere dem5 gop5 black latino female age education if choice_set == 11
test dem5=gop5
outreg2 using sitable2, word dec(2) append

* Figure 5: Gender Gap by Age *
* Under 40 *
reg vax_accept i.efficacy_n##i.female i.duration_n##i.female i.majorside_n##i.female i.minorside_n##i.female i.fda_n##i.female i.origin_n##i.female i.endorsed_n##i.female if age<40 , cluster(respondent) coeflegend
margins efficacy_n duration_n majorside_n minorside_n  fda_n origin_n endorsed_n, at(female=1) post
estimates store female
reg vax_accept i.efficacy_n##i.female i.duration_n##i.female i.majorside_n##i.female i.minorside_n##i.female i.fda_n##i.female i.origin_n##i.female i.endorsed_n##i.female if age<40, cluster(respondent) coeflegend
margins efficacy_n duration_n majorside_n minorside_n  fda_n origin_n endorsed_n, at(female=0) post
estimates store whites
coefplot  whites female, drop(_cons) omitted baselevels  ///
headings(1.efficacy_n = "{bf:Efficacy}" ///  
1.duration_n = "{bf:Protection Duration}" ///            
1.majorside_n = "{bf:Major Side Effects}" ///
1.minorside_n = "{bf:Minor Side Effects}" ///
1.fda_n = "{bf:FDA Approval}" ///
1.origin_n = "{bf:Origin}" ///
1.endorsed_n = "{bf:Endorsement}" , labsize(vsmall)) graphregion(color(white)) ylab(, labs(vsmall)) ///
xti("Marginal Mean of Taking Vaccine", size(small)) ti("Under 40") level(95) xlabel(.3 .4 .5 .6 .7 .8)  mlcolor(female) mcolor(female) msize(medsmall) legend(order(4 "Women" 2 "Men") size(small)) saving(under40.gph, replace)
gr_edit plotregion1.plot4.style.editstyle marker(symbol(square)) editcopy
* 40+ *
reg vax_accept i.efficacy_n##i.female i.duration_n##i.female i.majorside_n##i.female i.minorside_n##i.female i.fda_n##i.female i.origin_n##i.female i.endorsed_n##i.female if age>=40,  cluster(respondent) coeflegend
margins efficacy_n duration_n majorside_n minorside_n  fda_n origin_n endorsed_n, at(female=1) post
estimates store female
reg vax_accept i.efficacy_n##i.female i.duration_n##i.female i.majorside_n##i.female i.minorside_n##i.female i.fda_n##i.female i.origin_n##i.female i.endorsed_n##i.female if age>=40, cluster(respondent) coeflegend
margins efficacy_n duration_n majorside_n minorside_n  fda_n origin_n endorsed_n, at(female=0) post
estimates store whites
coefplot  whites female, drop(_cons) omitted baselevels  ///
headings(1.efficacy_n = "{bf:Efficacy}" ///  
1.duration_n = "{bf:Protection Duration}" ///            
1.majorside_n = "{bf:Major Side Effects}" ///
1.minorside_n = "{bf:Minor Side Effects}" ///
1.fda_n = "{bf:FDA Approval}" ///
1.origin_n = "{bf:Origin}" ///
1.endorsed_n = "{bf:Endorsement}" , labsize(vsmall)) graphregion(color(white)) ylab(, labs(vsmall)) ///
xti("Marginal Mean of Taking Vaccine", size(small)) ti("40+") level(95) xlabel(.3 .4 .5 .6 .7 .8)  mlcolor(female) mcolor(female) msize(medsmall) legend(order(4 "Women" 2 "Men") size(small)) saving(fortyplus.gph, replace)
gr_edit plotregion1.plot4.style.editstyle marker(symbol(square)) editcopy

grc1leg2  "under40.gph" "fortyplus.gph", rows(1) cols(2)
gr_edit style.editstyle boxstyle(shadestyle(color(white))) editcopy
gr_edit plotregion1.graph1.plotregion1.plot4.style.editstyle marker(symbol(Sh)) editcopy
gr_edit plotregion1.graph2.plotregion1.plot4.style.editstyle marker(symbol(Sh)) editcopy
gr_edit legend.plotregion1.key[1].view.style.editstyle marker(symbol(Sh)) editcopy


* SI Table 6 *
reg vax_accept i.efficacy_n i.duration_n i.majorside_n i.minorside_n i.fda_n i.origin_n i.endorsed_n i.age_group, robust cluster(respondent)
outreg2 using sitable6, word replace dec(2) label

